对链表进行排序 [英] Sorting a linked list

查看:31
本文介绍了对链表进行排序的处理方法,对大家解决问题具有一定的参考价值,需要的朋友们下面随着小编来一起学习吧!

问题描述

我用 C# 编写了一个基本的链表类.它有一个 Node 对象,它(显然)代表列表中的每个节点.

I have written a basic linked list class in C#. It has a Node object, which (obviously) represents every node in the list.

代码没有使用IEnumerable,但是可以实现排序功能吗?我使用的语言是 C#.在 C# 中有这样的例子吗?

The code does not use IEnumerable, however, can I implement a sorting function? The language I am using is C#. Is there an example of this in C#?

我正在使用这个示例:

谢谢

推荐答案

Functional Quicksort and Mergesort

这是一个以函数式风格编写的带有快速排序和归并排序方法的链表:

Functional Quicksort and Mergesort

Here's a linked list with quicksort and mergesort methods written in a functional style:

class List
{
    public int item;
    public List rest;

    public List(int item, List rest)
    {
        this.item = item;
        this.rest = rest;
    }

    // helper methods for quicksort

    public static List Append(List xs, List ys)
    {
        if (xs == null) return ys;
        else return new List(xs.item, Append(xs.rest, ys));
    }

    public static List Filter(Func<int,bool> p, List xs)
    {
        if (xs == null) return null;
        else if (p(xs.item)) return new List(xs.item, Filter(p, xs.rest));
        else return Filter(p, xs.rest);
    }

    public static List QSort(List xs)
    {
        if (xs == null) return null;
        else
        {
            int pivot = xs.item;
            List less = QSort(Filter(x => x <= pivot, xs.rest));
            List more = QSort(Filter(x => x > pivot, xs.rest));
            return Append(less, new List(pivot, more));
        }
    }

    // Helper methods for mergesort

    public static int Length(List xs)
    {
        if (xs == null) return 0;
        else return 1 + Length(xs.rest);
    }

    public static List Take(int n, List xs)
    {
        if (n == 0) return null;
        else return new List(xs.item, Take(n - 1, xs.rest));
    }

    public static List Drop(int n, List xs)
    {
        if (n == 0) return xs;
        else return Drop(n - 1, xs.rest);
    }

    public static List Merge(List xs, List ys)
    {
        if (xs == null) return ys;
        else if (ys == null) return xs;
        else if (xs.item <= ys.item) return new List(xs.item, Merge(xs.rest, ys));
        else return new List(ys.item, Merge(xs, ys.rest));
    }

    public static List MSort(List xs)
    {
        if (Length(xs) <= 1) return xs;
        else
        {
            int len = Length(xs) / 2;
            List left  = MSort(Take(len, xs));
            List right = MSort(Drop(len, xs));
            return Merge(left, right);
        }
    }

    public static string Show(List xs)
    {
        if(xs == null) return "";
        else return xs.item.ToString() + " " + Show(xs.rest);
    }
}

使用配对堆的函数式堆排序

奖励:堆排序(使用功能配对堆).

Functional heapsort using a Pairing Heap

Bonus: heapsort (using functional pairing heap).

class List
{
    // ...

    public static Heap List2Heap(List xs)
    {
        if (xs == null) return null;
        else return Heap.Merge(new Heap(null, xs.item, null), List2Heap(xs.rest));
    }

    public static List HSort(List xs)
    {
        return Heap.Heap2List(List2Heap(xs));
    }
}

class Heap
{
    Heap left;
    int min;
    Heap right;

    public Heap(Heap left, int min, Heap right)
    {
        this.left = left;
        this.min = min;
        this.right = right;
    }

    public static Heap Merge(Heap a, Heap b)
    {
        if (a == null) return b;
        if (b == null) return a;

        Heap smaller = a.min <= b.min ? a : b;
        Heap larger = a.min <= b.min ? b : a;
        return new Heap(smaller.left, smaller.min, Merge(smaller.right, larger));
    }

    public static Heap DeleteMin(Heap a)
    {
        return Merge(a.left, a.right);
    }

    public static List Heap2List(Heap a)
    {
        if (a == null) return null;
        else return new List(a.min, Heap2List(DeleteMin(a)));
    }
}

在实际使用中,您希望在不使用递归的情况下重写辅助方法,并且可能使用像内置列表那样的可变列表.

For actual use you want to rewrite the helper methods without using recursion, and maybe use a mutable list like the built-in one.

使用方法:

List xs = new List(4, new List(2, new List(3, new List(1, null))));
Console.WriteLine(List.Show(List.QSort(xs)));
Console.WriteLine(List.Show(List.MSort(xs)));
Console.WriteLine(List.Show(List.HSort(xs)));

链表的命令式就地快速排序

已请求就地版本.这是一个非常快速的实现.我从上到下编写了这段代码,没有寻找机会使代码变得更好,即每一行都是想到的第一行.非常难看,因为我使用null作为空列表:)缩进不一致等

Imperative In-place Quicksort for linked lists

An in-place version was requested. Here's a very quick implementation. I wrote this code top to bottom without looking for opportunities to make the code better, i.e. every line is the first line that came to mind. It's extremely ugly because I used null as the empty list :) The indentation is inconsistent, etc.

此外,我仅在一个示例上测试了此代码:

Additionally I tested this code on only one example:

        MList ys = new MList(4, new MList(2, new MList(3, new MList(1, null))));
        MList.QSortInPlace(ref ys);
        Console.WriteLine(MList.Show(ys));

神奇的是它第一次就成功了!我很确定这段代码包含错误.不要让我负责.

Magically it worked the first time! I'm pretty sure that this code contains bugs though. Don't hold me accountable.

class MList
{
    public int item;
    public MList rest;

    public MList(int item, MList rest)
    {
        this.item = item;
        this.rest = rest;
    }

    public static void QSortInPlace(ref MList xs)
    {
        if (xs == null) return;

        int pivot = xs.item;
        MList pivotNode = xs;
        xs = xs.rest;
        pivotNode.rest = null;
        // partition the list into two parts
        MList smaller = null; // items smaller than pivot
        MList larger = null; // items larger than pivot
        while (xs != null)
        {
            var rest = xs.rest;
            if (xs.item < pivot) {
                xs.rest = smaller;
                smaller = xs;
            } else {
                xs.rest = larger;
                larger = xs;
            }
            xs = rest;
        }

        // sort the smaller and larger lists
        QSortInPlace(ref smaller);
        QSortInPlace(ref larger);

        // append smaller + [pivot] + larger
        AppendInPlace(ref pivotNode, larger);
        AppendInPlace(ref smaller, pivotNode);
        xs = smaller;
    }

    public static void AppendInPlace(ref MList xs, MList ys)
    {
        if(xs == null){
            xs = ys;
            return;
        }

        // find the last node in xs
        MList last = xs;
        while (last.rest != null)
        {
            last = last.rest;
        }
        last.rest = ys;
    }

    public static string Show(MList xs)
    {
        if (xs == null) return "";
        else return xs.item.ToString() + " " + Show(xs.rest);
    }
}

这篇关于对链表进行排序的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持IT屋!

查看全文
登录 关闭
扫码关注1秒登录
发送“验证码”获取 | 15天全站免登陆